Developmentalists from the behavior genetic tradition have recently suggested that children evoke and actively select certain experiences, including parental experience, through the process of genotype-environment correlation. The primary objective of this project is to determine to what extent and how children create their own language environments. There is currently little empirical data with which to answer these important questions. Three studies are proposed. Two concern the bi-direction of effects in mother-child interaction (the longitudinal study and the dyad switch study.) These studies examine to what extent children create language environments by comparing the magnitude of mother effects with child effects. The third study concerns within-family environmental effects (the sibling study). This study examines how children create language environments by comparing the magnitude of individual difference effects in language production, verbal intelligence, and temperament. There are three major hypotheses. In the longitudinal study it is hypothesized that mothers' language at time 1 will predict children's language at time 2, that children's language at time 1 will predict mothers' language at time 2, and that child effects will be of similar magnitude to mother effects. In the dyad switch study it is hypothesized that mothers will adapt to individual differences between child partners more than children will adapt to individual differences between adult partners. In the sibling study it is hypothesized that mothers' language production with their offspring will be related to individual differences in children's language production, verbal intelligence, and temperament. If children do play a large role in making their own language environments, then the importance of genotype-environment correlation for understanding the role of experience in development will be documented. Practical implications also follow from these studies, which address the question of how language is learned, concerning sound, empirically-based intervention efforts.